September 30, 2012
Reforming the Patent System Toward a Minimalist System
Reforming the Patent System Toward a Minimalist System-Becker
The patent system (I concentrate on patents) is supposed to encourage innovation, the foundation of economic progress. It does this by giving patent recipients exclusive control for a specified period of time over use of their patented goods, machinery, processes, or services. It has long been recognized that patents impose costs on society since patents keep out competition, so that the monopoly power of patent holders enables them to raise prices and lower outputs. However, until recent years many other costs of the patent system received little attention, including paradoxically that this system might in fact discourage innovations.
Posner has a very good discussion of many costs of a patent system. These include not only the monopoly power given to patent recipients, but also time-consuming and expensive patent litigation because of overlapping patents, patent trolling to gain control of large numbers of patents not for use but to force compensation through litigation, and the ability to patent minor improvements over existing technologies. One additional important cost is that companies in many developing countries, in particular China, do not worry about patent infringement, and often produce much cheaper copies that undercut patented products. These companies have an advantage over companies in developed countries that refrain from producing patented products because they fear punishment from the courts for infringing patent rights.
Patent infringement lawsuits, copying by companies in developing countries, and many other obstacles to profiting from innovations under the modern patent system not only discourage certain types of patent applications, but may also reduce the incentive to innovate itself. Why apply for a patent if the information in the application makes it easier to produce copies? Moreover, the potential profit from an innovation might be more than dissipated through lengthy patent litigation that involves also the risk of large fines- as in the recently imposed fine on Samsung of over $1 billion for allegedly copying software patented by Apple.
The various harmful effects of the patent and copyright systems encouraged Arnold Plant, an English economist, to publish over 75 years ago two influential articles on why England and other countries would be better off without patents and copyrights. Among other things, he argued that the temporary monopoly power given to patent holders often led to inefficiently high prices and reduced output of patented goods, that many persons would continue to invent even if they could not patent their inventions, and that patents distort innovations in favor of goods and processes that can be patented and away from innovations that cannot be patented. His favorite example of the latter is basic research in the sciences that produced the theory of relativity, the theory of evolution, and in more modern times our understanding of DNA and genes. He believed that the patent system induced some creative scientists to work in areas that could be patented rather than on basic scientific research that could not be patented.
Although ending the patent system is a clean solution to all the problems induced by modern patenting, it clearly is not desirable given the importance of industries like the pharmaceutical industry. Since this industry spends on average hundreds of millions of dollars bringing to market a successful drug, pharmaceutical companies would not invest such large sums without the protection of patents (or without other benefits). Probably the best solution would be to maintain the patent system on drugs and a few other products that are expensive to innovate and cheap to copy, and eliminate patents on everything else. In particular, this means eliminating patents in the software industry, the source of much of the patent litigation and patent trolling.
I admit it is not clear where to draw the line between what should and should not be patentable. However, one can start by eliminating the ability to patent software, and go on from there to prune the number and type of inventions and innovations that are eligible for patent protection. Plant was not right about the desirability of eliminating the patent system, but he erred in the right direction. We should try to move much closer to where he wanted to be, and away from the present system.
Do patent and copyright law restrict competition and creativity excessively? Posner
I am concerned that both patent and copyright protection, though particularly the former, may be excessive.
To evaluate optimal patent protection for an invention, one has to consider both the cost of inventing and the cost of copying; the higher the ratio of the former to the latter, the greater the optimal patent protection for the inventor. The ratio is very high for pharmaceutical drugs. The cost of inventing a new drug, a cost that includes the extensive testing required for the drug to be approved for sale, is in the hundreds of millions of dollars, yet for most drugs the cost of copying—or producing an identical substitute—is very low. And so the ratio of the first to the second cost is very high, making it hard for the inventor to recover his costs without patent protection (and for the additional reasons that the present value of the revenue from sale of the drug is depressed because of the length of time it takes to get approval, and that the effective patent term is truncated because the patent is granted, and the period patent protection begins to run, when the patent is granted rather than, years later, when the drug can begin to be sold).
Pharmaceutical drugs are the poster child for patent protection. Few other products have the characteristics that make patent protection indispensable to the pharmaceutical industry. Most inventions are inexpensive, and even without patent protection, or any other legal protection from competition, the first firm to invent a product usually has significant protection from competition in the near term. The first firm gets a headstart on moving down his cost curve as experience demonstrates ways of cutting costs and improving the product. And the public is likely to identify his brand with the product, and keep buying it even after there is competition, and at a premium price. Moreover, many new products have only a short expected life, so that having 20 years of patent protection would confer no real benefit—except to enable the producer to extract license fees from firms wanting to make a different product that incorporates his invention.
When patent protection provides an inventor with more insulation from competition than he needed to have an adequate incentive to make the invention, the result is to increase market prices above efficient levels, causing distortions in the allocation of resources; to engender wasteful patent races—wasteful because of duplication of effort and because unnecessary to induce invention (though the races do increase the pace of invention); to increase the cost of searching the records of the Patent and Trademark Office in order to make sure one isn't going to be infinging someone's patent with your invention; to encourage the filing of defensive patents (because of anticipation that someone else will patent a similar product and accuse you of infringement); and to encourage patent "trolls," who buy up large numbers of patents for the sole purpose of extracting licensee fees by threat of suit, and if necessary sue, for infringement.
The problem of excessive patent protection is at present best illustrated by the software industry. This is a progressive, dynamic industry rife with invention. But the conditions that make patent protection essential in the pharmaceutical industry are absent. Nowadays most software innovation is incremental, created by teams of software engineers at modest cost, and also ephemeral—most software inventions are quickly superseded. Software innovation tends to be piecemeal—not entire devices, but components, so that a software device (a cellphone, a tablet, a laptop, etc.) may have tens of thousands, even hundreds of thousands, of separate components (bits of software code or bits of hardware), each one arguably patentable. The result is huge patent thickets, creating rich opportunities for trying to hamstring competitors by suing for infringement—and also for infringing, and then challenging the validity of the patent when the patentee sues you.
Further impediments to effective patent policy in the software industry include a shortage of patent examiners with the requisite technical skills, the limited technical competence of judges and jurors, the difficulty of assessing damages for infringement of a component rather than a complete product, and the instability of the software industry because of its technological dynamism, which creates incentives both to patent and to infringe patents and thus increases legal costs.
The pharmaceutical and software industries are the extremes so far as the social benefits and costs of patent protection are concerned, and there are many industries in between. My general sense, however, bolstered by an extensive academic literature, is that patent protection is on the whole excessive and that major reforms are necessary.
Turning to copyright, I note first an interesting contrast with patent law. Although there are some industry-specific differences in patent law, for the most part patents are "one size fits all," so far as length of protection and criteria and procedures for the grant of a patent are concerned. In contrast, copyright protection tends to vary considerably across different media. For example, when recorded music came into being, instead of providing it with the same copyright regime as already governed books and other printed material, Congress devised a separate regime tailored to what were considered the distinctive characteristics of music as a form of intellectual property. Patent law could learn from that approach.
The problem of copyright law is less acute than the problem of patent law, partly because copyright infringement is limited to deliberate copying; patent infringement does not require proof even that the infringer was aware of the patent that he was infringing. Nevertheless, as in the case of patent law, copyright protection seems on the whole too extensive. Granted, with modern action movies often costing hundreds of millions of dollars to make, yet copiable almost instantanteously and able to be both copied and distributed almost costlessly, the need for copyright protection is comparable to that in the pharmaceutical industry. At the other extreme is academic books and articles (apart from textbooks), which are produced as a byproduct of academic research that the author must conduct in order to preserve his professional reputation and that would continue to be produced even if not copyrightable at all. It is doubtful that there is any social benefit to the copyrighting of academic work other than textbooks, which require a lot of work and generally do not enhance the author's academic reputation and may undermine it.
The most serious problem with copyright law is the length of copyright protection, which for most works is now from the creation of the work to 70 years after the author's death. Apart from the fact that the present value of income received so far in the future is negligible, obtaining copyright licenses on very old works is difficult because not only is the author in all likelihood dead, but his heirs or other owners of the copyright may be difficult or even impossible to identify or find. The copyright term should be shorter.
The next most serious problem is the courts' narrow interpretation of "fair use." The fair use defense to copyright infringement permits the copying of short excerpts from a copyrighted work without a license, since the transaction costs of negotiating a license for a short excerpt would tend to exceed the value of the license. The problem is that the boundaries of fair use are ill defined, and copyright owners try to narrow them as much as possible, insisting for example that even minute excerpts from a film cannot be reproduced without a license. Intellectual creativity in fact if not in legend is rarely a matter of creation ex nihilo; it is much more often incremental improvement on existing, often copyrighted, work, so that a narrow interpretation of fair use can have very damaging effects on creativity. This is not widely recognized.
The need for reform is less acute in copyright than in patent law, but it is sufficiently acute to warrant serious attention from Congress and the courts.